It has for some time been established that infective juveniles of the order Rhabditida, especially those of the genera Steinernema and Heterorhabditida are able to effect killing of the German cockroach Blatella germanica as well as other insects. In tests using direct application of the nematodes to subject cockroaches with a small sponge soaked in nematodes or a petri dish lined with filter paper containing the nematodes, high levels of mortality were obtained in cockroaches using infective juveniles from these genera, especially the species Steinernema feltiae previously known as Neoaplactana carpocapsae. A number of groups have reported these results. See, for example, Locatelli, D. P. et al., La Difesa delle Piante (1987) 10:339-348; Zukowski, K., Roczniki Panstwowego Zakladu Higieny (1984) 35:451-457; Boush, G. M., University of Wisconsin, Madison, Wis., Reports of Results of Government Contracts, 18326 (1983); 18226 (1982); 18206 (1981); and Skierska, B. et al., Bull Inst Maritime Trop Med (1976) 27:207-227. The ability of H. heliothidis to kill the American species Periplaneta americana has also been confirmed by Zervos, S. et al , Can J Zool (1989) 67:1609-1611.
However, mere knowledge that certain species of entomogenous nematodes can successfully kill insects when the infective juveniles are directly introduced to them does not permit successful design of traps which will capitalize on this ability of the nematodes in a practical environment. Advantage must be taken of the behavior of the target insect so that the insect is brought into proximity with the infective juveniles (IJs) and the IJs themselves must be maintained in a condition to effect the infestation of the insects. The design of a successful roach trap, fly trap, yellow jacket trap, or trap for other insects therefore, requires an understanding of the behavior of both the target insect and the infectious agent, as well as a plan for construction which takes advantage of these characteristics.
The need for providing a response to at least some of these parameters in an analogous context is recognized in British patent GB 2,174,907, granted 19 Oct. 1988. In the apparatus described, which is designed specifically for flying insects, live entomophilic nematodes are maintained on a moist absorbent pad which is kept from desiccating by a supply of liquid water either directly added to the pad through an external source or conducted into the pad by capillary action from a reservoir of liquid water with which the pad is in contact. The description further indicates that an attractant for the insect, such as a pheromone, must be included in order to provide incentive for the flying insects to contact the pad containing the nematodes. All of the designs are open to the atmosphere, as the aim is to attract flying insects specifically. Although it is mentioned in the description that these apparatuses may be used to control cockroaches, it is clear that the designs presented are not appropriate to do so since they would require the cockroach to surmount vertical surfaces--a behavior not attractive to these insects. The described apparatuses have other disadvantages as well--as the nematodes are evenly distributed throughout the absorbent pad, and as the pad is kept saturated with liquid water, the nematodes do not exhibit nictation behavior, and approximately 70% of them escape from the trap within a day. Furthermore, practical quantities of liquid water can keep the absorbent pad moist for only 3-7 days without adding more water to the reservoir.
It would be useful to provide a device to control roaches and other insects in an environment which could be left in place for substantial periods of time, which does not, in the case of roaches at least, require expensive attractants or bait and which does not pose a hazard to unintended targets. The present invention provides such a device.